INDEX

Explanations

phrases related to rights and justification

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

arel

-0.16

ral

-0.15

zaw

-0.14

ìľ¡

-0.13

asser

-0.13

ãĥ¼ãĥĲ

-0.13

ahkan

-0.12

oola

-0.12

awk

-0.12

zar

-0.12

POSITIVE LOGITS

 reason

1.01

reason

0.77

 reasons

0.74

Reason

0.72

 Reason

0.71

 Reasons

0.62

 cause

0.62

çĲĨçĶ±

0.60

.reason

0.58

_reason

0.57

Activations Density 0.201%

phrases related to rights and justification

No Comments

No Known Activations